Professional Documents
Culture Documents
Learning Objectives After completing this session you should be able to explain what a work process is and why it is so describe a systematic approach for improving business processes discuss common pitfalls encountered when working on a process improvement project use basic tools for improving process
Suggested reading: The Golden Rules of Process Redesign by Roger Rupp and James Russell
Session Agenda
Process introduction or understanding process Your examples on process Process improvement - TQM Your examples on improvement Improvement process models and pitfalls Your ideas on process Tools for quality management Your experience on tools
Process Definition
Thomas Davenport, in Process Innovation: ". . . a structured, measured set of activities designed to produce a specified output for a particular customer or market." Michael Hammer and James Champy, in Reengineering the Corporation: ". . . a collection of activities that takes one or more kinds of input and creates an output that is of value to the customer."
Process Definition
James Harrington, in Business Process Improvement "Any activity or group of activities that takes an input, adds value to it, and provides an output to an internal or external customer." Joseph M. Juran, in Juran on Planning for Quality ". . . a systematic series of actions directed to the achievement of a goal."
Four groups of people are involved in the operation and improvement of processes: Customers Work Group Supplier Owner
Why process?
We are quality talking; then conformance, variation, gapsbetween gapsbetween customer expectation and what we offer. A customer focuses on value they get (along the visible line when you have to sketch service blueprint last term), he does not care of each business function. Colleagues as customers Responsiveness to customer inquiries. then...
Why process?
with process and detailed elements we have: sequence (quality) who produce the outputs (accountability) boundary clarification (responsibility and quality) process owner (accountability) input & output those help promote quality
Production process
Inputs Process
- Assembly
Outputs
- Customer receives product
- Component
parts
- machines - Operators
Feedback
Information on product performance
Process
- Synthesize
Outputs
Employee to fill vacant position
Feedback
Inform personnel decision-maker of strengths and weaknesses of employee to improve future hiring efforts
Identifying processes Why difficult? We are talking more on function We have no chance to get out of out functional department It is easier to discuss on function, task, step, or operation. Its hard to talk about process in a business Functions have names, processes have no names. Disputes often arise when talking process. Process hides in stream of activities.
Identifying process How to? Anchor the analysis of a business process to something, some one, tangible, or customer. Ignore process name Set boundaries anywhere on stream to analyse (tips: where state changes, transfer of ownership, hand-offs) hand beware of loop in process.
Input Output Process identifying Anchors setting Get related people involved Problem Identifying tools Process improvement
The boundaries of a process may not be limited in a function Intrafunctional process is a process performed within a single function (e.g. preparation of wage sheets) Interfunctional process is a process involving more than one function (e.g. product development and planning)
1. Which products and services are most important to the customers? (implication-priority, general (implicationview of customer requirement) 2. What are the processes that produce these products and services? (general view) 3. What are the key ingredients that stimulate action in the organization, and what are the processes that convert these stimuli to output? (more detailed of process elements)
4. Which processes have the highest visibility with customers? (priority) 5. Which processes have the greatest impact on customercustomer-driven performance standards? (key process identified) 6. Which processes do performance data or common sense suggest have the greatest potential for improvement? (improvement opportunities)
Continuous improvement is a systematic approach to the closing of gaps between customer expectations and the characteristics of process outputs.
Histograms
Check Sheets
Pareto Diagrams
Scatter Diagrams
1. FLOWCHARTS
What is a flowchart?
A flowchart is a pictorial summary of the flows and decisions that comprise a process
Flowchart Symbols
The general form that represents input or output media, operations, or processes is a parallelogram
Decision symbol
A diamond is the symbol that denotes a decision point in the process. It includes attribute type decision such as pass fail, yes no. It also includes variable type decisions such as which of several categories a process measurement falls into. A line with an arrowhead is the symbol that shows the direction of the stages in a process. The flowline connects the elements of the system. The general symbol used to indicate the beginning and end of a process is an oval.
Flowline symbol
Types of Flowcharts
Systems Flowchart
A system flowchart is a pictorial summary of the sequence of operations that make up a process. It shows whats being done in a process. Applications: manufacturing process, quality of design study, quality control chart, organization chart, flows of billing, accounting, purchasing, etc.
Bad
Bad
START
STOP
Layout Flowchart
A layout flowchart depicts the floor plan of an area, usually including the flow of paperwork or goods and location of equipment, file cabinets, storage areas, and so on. These flowcharts are especially helpful in improving the layout to more efficiently utilize a space.
Rule 1: The most important rule is that the right 1: people must be involved in making the chart. Rule 2: All members of the group must participate. 2: The use of an independent facilitator will be a great benefit. Rule 3: All data must be visible and 3: understandable to all the people all the time.
Rule 4: Enough time must be allotted. 4: Rule 5: The more questions everyone asks, the 5: better. Questions are the key to the flowcharting process. What is the first thing that happens? What is the next thing that happens?
Benefits of a flowchart
It functions as a communication tool It functions as a planning tool It provides an overview of the system It defines the roles It demonstrates interrelationships It promotes logical accuracy It facilitates trouble-shooting troubleIt documents a system
The technique consists of defining an occurrence or problem (effect). Once the effect is defined, the factors that contribute to it (causes) are delineated Relationship between causes and effects
Quality Problem
Environment Materials Method
3. Brainstorm subcauses
Personnel Knowledge Absenteeism Machine Measurement Repeatability Capacity Sample size
Capability Safety
Quality Problem
Temperature Light Quality Working Environment Materials Effectiveness Method Standardization
Advantages: Forces preparation of the process flow diagram Forces consideration of the entire process as potential causes of a problem Identifies alternative work procedures Educational for members not familiar with the entire process Easy to use since most group members will be familiar with the process Can be used to anticipate process problems by focusing on sources of variability.
Disadvantages Easy to overlook potential causes (e.g., materials or measurements) since people may be too familiar with the process. Hard to use on long, complicated processes
Investigation of Potential Causes The C&E diagram is not intended to identify which potential causes are root causes. Thus, it is necessary to identify whether a potential cause is a root cause of a problem
C&E diagram
Prioritization
Hypothesis generation
Design studies
C&E diagram
Select most likely potential causes. Evaluate each against problem specification
Analyze data
No
Yes No
Problem eliminated Correct problems and monitor results
Yes
Stop Process for identifying whether a potential cause is a root cause of a problem
3. CONTROL CHARTS The major kinds of control charts: Attribute Data: Charts for nonconforming units ( p chart) Charts for counts of nonconformities (c chart) Variable Data: R chart X-bar chart Process Capability refers to the inherent variability of process output relative to the variation allowed by the design specifications.
4. HISTOGRAMS
Histograms are a graphic display of the number of times a measurement has occurred at a particular value or within an interval of the measurement scale. scale.
Bell-shaped
Double peaked
Plateau
(a)
(b)
Skewed
Truncated
Isolated peaked
Edge peaked
Constructing a Histogram
a. Quick Method A histogram can be constructed using the following steps: 1. Find the largest and smallest measurement 2. Construct the measurement scale between the largest and smallest measurement
Smallest measurement Largest measurement
3. Place a slash \ on the scale in step 2 for each occurrence \ of a measurement. Every fifth occurrence should use a longer slash ( \ \ \ \ ) . A space should be skipped between group of five
4. Evaluate whether there are too many intervals in the main body of the histogram. If there are too many intervals, the interval method should be used.
b. Interval Method A histogram can be constructed using the following steps: 1. Find the largest and smallest measurement 2. Choose a value for the number of intervals using the preceding guidelines for the number of measurements to be plotted in the histogram 3. Compute the interval size:
l ! interval size ! ( largest measuremen t - smallest measuremen t ) number of scale intervals
4. Round l to a convenient number. For example, number. 0.512 becomes 0.5 5. Choose a convenient number slightly smaller than or equal to the smallest measurement; call it measurement; Xlow. Compute the following intervals: intervals: low. Xlow Interval 1 Xlow + l
Xlow + 2l Xlow + 3l
Interval 2 Interval 3
6. Construct a scale to be used in plotting that corresponds to the intervals computed in step 5:
Smallest measurement Largest measurement
Xlow Xlow + 1l
Xlow + 2l Xlow + 3l
7. Place a slash \ in the interval in step 6 for each occurrence of a measurement. measurement.
Xlow
Xlow + 1l
Xlow + 2l
Xlow + 3l
For either the quick or the interval method, shaded bars are often drawn around the slashes. A slashes. frequency (number of occurrences) scale is typically used. used.
Frequency
Xlow
Xlow + 1l
Xlow + 2l
Xlow + 3l
Histogram from Control Charts There are three types of histograms that can be obtained form typical X-bar and R chart data: data:
Sample 1 2 3 N 1 Subgroup (n = 5) 2 3 4 5 X R Individuals histogram
5. CHECK SHEET
Check sheets are used for collecting or gathering data in a logical format. The most important purpose of check sheet is to enable the user(s) to gather and organize data in a format that permits efficient, easy analysis.
Attribute check sheet. Gathering data about sheet. defects in a process is necessary for stabilization, improvement, and innovations of the process. Variables check sheet. Gathering data about sheet. a process also involves collecting information about variables, such as size, length, weight, temperature, etc. These data are best etc. represented by organizing the measurements into a frequency distribution on a variables check sheet. sheet.
Defect location check sheet. Another way to sheet. gather information about defect in a product is to be used a defect location check sheet. This is a picture of a product or a portion of it on which an inspector indicates the location and nature of the defect. Checklist. Checklist. Checklist is used under operational conditions to ensure that all important steps or action have been taken. Its purpose is for guiding taken. operations, not for collecting data. The preflight data. checklist that a commercial airline pilot must complete is a good example. example.
S S S S
S S
8 Are all stratification variables identified, such as machine, spindle, pallet and fixture? Are such factors as vendor, test hand, furnace, die, tool and material identified? Are any lot identification numbers records? 9 9Will total number of parts evaluated be recorded (how many)? Is the number of parts produced during the evaluation period indicated? 10 Do individuals understand how to record multiple defects? Will all defects on a part be evaluated? Do individuals understand how multiple occurrences of the same defect on a unit will be recorded (e.g., is 1 missing screw or 10 missing screws on a unit recorded as a single defect)? 11 Are data recorded on check sheets in a manner analogous to the gage measurement scales? 12 Is coded gage data used if appropriate? Is the relationship between the coded gage data and the actual measurement state as: Actual value = offset + scale x gage value
S S
13. Has a trial run of the check sheet been attempted? Has a trial analysis of the data been attempted? 14. Is there sufficient space for the user of the check sheet to write in and make comments concerning unusual occurrences?
6. PARETO DIAGRAM
Pareto focused attention on the concept of the vital few versus the trivial many. Pareto is a tool used to identify and prioritize problems for solution. The 80-20 rule is based on an application of 80the Pareto principle. For example, 80 percent of sales come from 20 percent of customers, or 80 percent of dollar inventory is accounted for by 20 percent of the items.
Constructing a Pareto Diagram 1. Establish categories for the data being analyzed. 2. Specify the time period during data will be collected. 3. Construct a frequency table arranging the categories from the one with the largest number of observations to the one with the smallest number of observations.
Category
Signal
Frequency
Relative %
(1) (2) (3) (4) 1. XX Y1 C1 C1/C x 100 = P1 2. XX Y2 C2 C2/C x 100 = P2 3. XX Y3 C3 C3/C x 100 = P3 Total C 100% N = Total number of parts or items inspected
C/N
4. Construct a Pareto diagram. Part A, Pareto bars: Select the most appropriate scale to plot the Pareto bars: Frequency = Column 3 Relative % = Column 4 % of total inspected = Column 6
Part B, cumulative percentages: Select the line plot or stacked bar format to graph the cumulative percentages and plot the data in column 5.
Defects Undersized A hole Porosity Nicks Concentricity diameter A-F Parallelism surface Z-W Oversized A hole Others
Total defects: Number inspected:
C = 557 N = 2217
Pareto Analysis Work Sheet Defects (1) Undersized Nicks Parallelism Porosity Concentricity Oversized Others Total Signal Frequency Relative Cumulative % % (2) (3) (4) (5) A 224 40 40 B 149 27 67 C 58 11 78 D 52 9 87 E 46 8 95 F 5 1 96 G 23 4 100 557 100% 100% % of Total Inspected (6) 10.1 6.7 2.6 2.3 2.1 0.2 1.0 557/2217 = 25.1
ed
s it y
ze
ck
it y
Un
Pa
Co
nc
Ov er
Po
de
ra
en
Ot he
li s
r si
Ni
ll e
tr ic
ro
s iz
rs
100
200
80
150
60
100
40
50
20
Cumulative percentage
150
100
50
Sometime Pareto diagram can have more impact when problems or defects are represented in terms of their dollar costs. costs.
Pareto Analysis Cost Work Sheets: Cost per Defect Work Sheet Signal (2) A B C D E F Frequency (3) 224 149 58 52 46 5 534 Cost per unit ($) (4) 0.25 0.10 1.25 0.75 1.25 1.25 Cost of defect (5) 56.00 14.90 72.50 39.00 57.50 6.25 246.15 Rank (6) 3 5 1 4 2 6
Pareto Analysis Cost Work Sheet Defect (1) Parallelism Concentricity Undersized Porosity Nicks Oversized Total Signal (2) C E A D B F Frequency (3) 58 46 224 52 149 5 534 Cost of defect (4) 72.50 57.50 56.00 39.00 14.90 6.25 246.15 Relative cost Cumulative % % (5) (6) 29 29 23 52 23 75 16 91 6 97 3 100
100 80 70 60 50 40 30 20 10 0 F C u m u la tiv e p e r c e n ta g e 90
y y y
y y
Ignore customers requirements Data collected from unstable processes Identify the easy-to-find problems as the easy-tovital few Ignore the trivial many problems Short time for Pareto analysis
Check Sheet
Collect Data
Instructions
Pareto diagram
Problem A
7. SCATTER DIAGRAM
Scatter diagram provides a method to evaluate these interrelationships between two characteristics Understanding the relationship between characteristics increases our ability to control a process and detect problems.
Types of Patterns
Characteristics 2 Characteristics 2 Characteristics 2
Characteristics 1
Characteristics 1
Characteristics 1
(a)
Characteristics 2
(b)
Characteristics 2
(c)
Characteristics 1
Characteristics 1
(d)
(e)
First, it is assumed that if correlation exists between two characteristics, one characteristic in some way causes the value of the other characteristic This is not true Correlation does not imply causation. The second mistake commonly made in using scatter plots is to consider too narrow a range of the characteristics being evaluated. evaluated.
It is a good practice to use a sample size of at least N = 30 50 sampling units. units. 1.Select the characteristic (label 1), if any, that for some reason is thought to predict the value of the other characteristic (label 2).
2. Plot the values for each sampling unit. If repeated unit. values occur in a particular location, use the following plotting convention: convention:
Single value Two values Three values Four values
3. Use the types of patterns above to evaluate the relationship between the characteristics. The next characteristics. two procedures can be used to test for the presence of correlation
Median 1 Characteristic 2
2 3
1 4
Median 2
Characteristic 1
4. Count the number of points in each quadrant, and compute the following quantities: A = number of points in quadrants 1 and 3 B = number of points in quadrants 2 and 4 Q = number of points on either median line Total N = A + B + Q
5. Compute the test value C, where C = smaller of A and B = min (A, B) Sample size for the test SS = N Q If C is no greater than the table value c for the given value of SS, the characteristics are correlated. The maximum value c of computed C needed to conclude that a correlation exists:
SS 20-22 23-24 25-27 28-29 30-32 33-34 35-36 37-39 40-41 42-43 44-46 47-48 49-50 51-53
c 5 6 7 8 9 10 11 12 13 14 15 16 17 18
SS 54-55 56-57 58-60 61-62 63-64 65-66 67-69 70-71 72-73 74-76 77-78 79-80 81-82 83-85
c 19 20 21 22 23 24 25 26 27 28 29 30 31 32
SS 86-87 88-89 90-91 92-93 94-96 97-98 99-101 110 120 130 140 150 200
c 33 34 35 36 37 38 39 43 48 52 57 61 84
The seven-step process improvement model sevenintroduces a systematic approach for applying quality management to any type of process. The approach helps build a fundamental understanding of the business processes before attempting to improve them. The seven steps are described in a straightforward, linear fashion for clarity. Actual application will often be complicated by the need to recycle back to earlier steps as new information is gained and earlier hypotheses are found to be incomplete or incorrect.
1 2 3 4 5 6 7
Identify the gap between what is and what should be Document the importance and relevance of the problem List the customers key quality characteristics State how closing the gap will benefit customers Identify the outputs Identify processes producing these outputs Identify the process owner
Develop flowchart of the processes Identify the process participants Determine what data to be used to measure progress Customer satisfaction Cost of quality Output delivered Process variables (what, where, to what extent and who)
Design data collection instrument, develop any operation definitions needed to collect data Collect data, summarize the information learned about variables effect on the problem
Use the data collected and the experience of the process participants to identify conditions that might lead to the problem Analyze the conditions of interest and decide on most likely causes If possible verify the cause through observation or by directly controlling variables
Develop a list of solutions to be considered Assess the feasibility of each solution, likelihood of success and potential adverse consequences Indicate clearly why a particular solution is chosen Plan the implementation (Any pilot project? Who is responsible? Any training required?) Implement the preferred solution
Collect data on measurements used to measure progress decided in step two Analyze the results. Determine if the solution tested was effective Repeat prior steps as necessary
Develop a plan for institutionalizing the improvement and assign responsibilities Implement the plan, check to see that it has been successful
Determine if the remaining gap should be narrow further Identify related problems that should be addressed Summarize lessons learned during the process improvement project, make recommendations for future teams Celebrate the contributions of the team.
7-step method problem solving for E-MBA is a Esystematic one for improvement rather than using the tools alone and intermittently. In these steps, using the 7 tools of improvement for finding causes and solutions Where and which tools to use?
There are other methods of problem solving but the tools are useful there as well.
P (Plan) A (Act)
D (Do) C (Check)
Step 4
Steps 6 & 7
Step 5
Quality
Plan-Do-CheckPlan-Do-Check-Act Procedure
Plan. Recognize an opportunity and plan a change. Do. Test the change. Carry out a small-scale smallstudy. Check. Review the test, analyze the results and identify what youve learned.
Plan-Do-CheckPlan-Do-Check-Act Procedure
Act. Take action based on what you learned in the study step: If the change did not work, go through the cycle again with a different plan. If you were successful, incorporate what you learned from the test into wider changes. Use what you learned to plan new improvements, beginning the cycle again.
As a model for continuous improvement. When starting a new improvement project. When developing a new or improved design of a process, product or service. When defining a repetitive work process. When planning data collection and analysis in order to verify and prioritize problems or root causes. When implementing any change.
The Philip Crosby Associates Method for eliminating Non-conformances Non1 Define Situation
Fix Situation
Corrective Action
4 Develop Specifications 5 Identify Process Steps 6 7 Select Measurements Determine Capability OK ? Produce Output 8 Evaluate Results OK ? 9 Recycle Problem Solve Problem Solve
Organizations have customized the process improvement model for their own use. Some modified the terminology. Some reduced the number of steps by combining elements, and others increased the number of steps to explicitly consider certain elements. These variants of the process improvement model are all somewhat different from one another but in essence they are all very similar.
Although numerous problems will be encountered along the seven steps, those included in the checklist impact the early steps and often cause irreparable damage. Tampering Incomplete ownership Lack of expertise Imposed solution Constraints World hunger
y y y y y y
Concluding remarks
Systematic process improvement relies on building a fundamental understanding of customers requirements, process capability and the causes for the gaps between them. Hypotheses are developed and tested and improvement gained through the continuous cycle of plan-do-check-act. This systematic approach plan-do-checkbears striking contrast to the classic short cut of problem detection and subsequent solving, an approach resembling plan-do-plan-do. plan-do-plan-
The seven-step model was developed as a sevenuniversal guide for systematic improvement. It can be used to improve manufacturing as well as nonnonmanufacturing processes and helps assure application of the plan-do-check-act strategy. plan-do-check-